School of Biomedical and Health Sciences, Victoria University, Melbourne, Australia.
J Altern Complement Med. 2012 Nov;18(11):1028-37. doi: 10.1089/acm.2011.0621. Epub 2012 Aug 16.
Chinese medicine (CM) has its own diagnostic indicators that are used as evidence of change in a patient's condition. The majority of studies investigating efficacy of Chinese herbal medicine (CHM) have utilized biomedical diagnostic endpoints. For CM clinical diagnostic variables to be incorporated into clinical trial designs, there would need to be evidence that these diagnostic variables are reliable. Previous studies have indicated that the reliability of CM syndrome diagnosis is variable. Little information is known about where the variability stems from--the basic data collection level or the synthesis of diagnostic data, or both. No previous studies have investigated systematically the reliability of all four diagnostic methods used in the CM diagnostic process (Inquiry, Inspection, Auscultation/Olfaction, and Palpation).
The objective of this study was to assess the inter-rater reliability of data collected using the four diagnostic methods of CM in Australian patients with knee osteoarthritis (OA), in order to investigate if CM variables could be used with confidence as diagnostic endpoints in a clinical trial investigating the efficacy of a CHM in treating OA.
An inter-rater reliability study was conducted as a substudy of a clinical trial investigating the treatment of knee OA with Chinese herbal medicine. Two (2) experienced CM practitioners conducted a CM examination separately, within 2 hours of each other, in 40 participants. A CM assessment form was utilized to record the diagnostic data. Cohen's κ coefficient was used as a measure of the level of agreement between 2 practitioners.
There was a relatively good level of agreement for Inquiry and Auscultation variables, and, in general, a low level of agreement for (visual) Inspection and Palpation variables.
There was variation in the level of agreement between 2 practitioners on clinical information collected using the Four Diagnostic Methods of a CM examination. Some aspects of CM diagnosis appear to be reliable, while others are not. Based on these results, it was inappropriate to use CM diagnostic variables as diagnostic endpoints in the main study, which was an investigation of efficacy of CHM treatment of knee OA.
中医（CM）有自己的诊断指标，用于作为患者病情变化的证据。大多数研究中药（CHM）疗效的研究都使用了生物医学诊断终点。要将中医临床诊断变量纳入临床试验设计，就需要有证据表明这些诊断变量是可靠的。以前的研究表明，中医证候诊断的可靠性是可变的。对于这种可变性的来源，人们知之甚少——是基本数据采集层面还是诊断数据的综合，或者两者兼而有之。以前没有研究系统地调查中医诊断过程中使用的四种诊断方法（问诊、望诊、闻诊和切诊）的可靠性。
本研究旨在评估澳大利亚膝骨关节炎（OA）患者使用中医四种诊断方法收集的数据的组内一致性，以调查中医变量是否可以作为临床试验中治疗 OA 的 CHM 疗效的诊断终点，从而有信心地使用这些变量。
这是一项中药治疗膝骨关节炎的临床试验的子研究，进行了一项组内一致性研究。两名（2 名）经验丰富的中医从业者在 2 小时内分别对 40 名参与者进行了中医检查。使用中医评估表记录诊断数据。使用 Cohen's κ 系数作为衡量两名从业者之间一致性水平的指标。
在问诊和闻诊变量方面，有相对较好的一致性，而在（视觉）望诊和切诊变量方面，总体上一致性较低。
两名从业者在使用中医四诊方法收集的临床信息方面存在一致性差异。中医诊断的某些方面似乎是可靠的，而其他方面则不可靠。基于这些结果，在主要研究中，将中医诊断变量作为 CHM 治疗膝骨关节炎疗效的诊断终点是不合适的。
